Research And Development On Monitoring Software And BOD Soft Measurement System For Wastewater Treatmeat Process

The basic goal of urban wastewater treatment process is to purify domestic and production wastewater, and to make sure that the quality of effluent water up to the national standard. Biochemical oxygen demand(BOD) is a parameter to reflect the purity of biodegradable organism in the water, which also is one of the key indicators of water quality evaluation. By using the traditional BOD measuring method, the final measurements will be got after five days, which involves a series of complicated biochemical reaction process including field sampling and laboratory training process. The hysteretic character turns out that real time monitoring and on-line adjustment by using BOD couldn’t be achieved, which makes it difficult to keep the stability. Therefore, the paper concentrates on research and application of monitoring software and BOD soft measurement system, combined with a control system development project in a domestic wastewater treatment plant. The main contents as following:(1) The development on Monitoring software. According to the engineering drawings, the hardware Configuration of automatic control system for wastewater treatment has been completed, and the logic control procedures of the running equipments have been written based on technological requirements. Then the operating condition of wastewater treatment process could be monitored in real-time on the basis of the design of wastewater treatment process and the configuration monitor screen. On the basis of these, the paper also designs a data report used to store the important parameters about water quality and variable changes in related process, which is a basis of analyzing the process conditions and real-time predicting BOD concentration.(2) The research and development on BOD soft measurement system. Because BOD concentration is difficult to real-time monitor, the paper establishes a case-based reasoning(CBR) soft measurement model based on the revised method of support vector regression machine(SVR). This model mainly includes case retrieval, case reuse, SVR revision and case storage. SVR revision model is established by using historical data to build revised case base, and SVR training, which could revise BOD concentration value got from the traditional CBR solving model. The result of the comparative experiment shows that the accuracy of prediction model by using SVR revised method is better than the results by using support vector machine(SVM), BP neural network, RBF neural network and traditional CBR method. And it also shows that SVR revised method could be used to improve CBR regression performance, reduce BOD prediction error, and meet the requirement of monitoring.(3) The research on engineering application. Based on BOD soft measurement model, the paper programs by using VBS language of upper computer configuration software, and designs the soft-measuring monitor screen. The results of the field debugging and experiment show that soft measurement system is running very well which could take place of the traditional manual test method by achieving BOD real-time monitoring, and is a foundation of achieving the stability control in the wastewater treatment process.